Input Ranking Revisited: A Theoretical Input Sensitivity Approach

  • Sanggil Kang
  • Can Isik
  • Steven Morphet
Conference paper
Part of the Advances in Soft Computing book series (AINSC, volume 19)


In feedforward neural networks, all inputs contribute to a greater or lesser extent when calculating the outputs. Therefore, inputs may be ordered from the greatest contributor to the least. Input ranking is non-trivial — cursory examination of the weight and bias matrices fails to reveal ranking. Solving the ranking issue allows for elimination of inputs with little influence on output. This paper presents a new method of determining the input sensitivity of three-layer feedforward neural networks. Specifically, sensitivity of an input is independent of the magnitudes of the remaining inputs, providing an unambiguous ranking of input importance. Small changes to influential inputs will result in great changes to output. This concept motivated the theoretical approach to input ranking. Examination of theoretical results will demonstrate the correctness of this approach.


Neural Network Hide Neuron Feedforward Neural Network Input Neuron Joint Probability Density Function 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Choi, J. Y.; Choi, C. H. Sensitivity Analysis of Multilayer Perceptron with Differentiable Activation Functions. IEEE Transactions on Neural Networks. vol. 3, no. 1, 1992, pp. 101–107.Google Scholar
  2. 2.
    Engelbrecht, A. P.; Cloete, I.; Zurada, J. M. Determining the Significance of Input Parameters Using Sensitivity Analysis. Proceedings 1995 International Workshop on Artificial Neural Networks. pp. 382–388.Google Scholar
  3. 3.
    He, F.; Sung, A. H.; Guo, B. A Neural Network Model for Prediction of Oil Well Cement Bonding Quality. Proceedings 1997 IASTED International Conference on Control. pp. 417–420.Google Scholar
  4. 4.
    Piche, S. Robustness of Feedforward Neural Networks. International Joint Conference on Neural Networks. vol. 2, 1992, pp. 346–351.Google Scholar
  5. 5.
    Sung, A. H. Ranking Input Importance in Neural Network Modeling of Engineering Problems. Proceedings 1998 IEEE International Joint Conference on Neural Networks. vol. 1, pp. 316–321.Google Scholar
  6. 6.
    Zurada, J. M.; Malinowski, A.; Cloete, I. Sensitivity Analysis for Minimization of Input Data Dimension for Feedforward Neural Network. Proceedings 1994 IEEE International Symposium on Circuits and Systems. vol. 6, pp. 447–450.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Sanggil Kang
    • 1
  • Can Isik
    • 1
  • Steven Morphet
    • 1
    • 2
  1. 1.Department of Electrical Engineering and Computer ScienceSyracuse UniversitySyracuseUSA
  2. 2.Syracuse Research CorporationNorth SyracuseUSA

Personalised recommendations